Updated on 2026/03/04

写真a

 
ONO ISAO
 
Organization
School of Computing Professor
Title
Professor
External link

Degree

  • Doctor ( Tokyo Institute of Technology )

Research Areas

  • Informatics / Soft computing

Education

  • Tokyo Institute of Technology   Interdisciplinary Science and Engineering

    - 1997

      More details

    Country: Japan

    researchmap

Research History

  • -:

    2005

      More details

  • -:東京工業大学 大学院総合理工学研究科 助教授

    2005

      More details

  • :

    2001 - 2005

      More details

  • :徳島大学 工学部 助教授

    2001 - 2005

      More details

  • :徳島大学 工学部 講師

    1998 - 2001

      More details

  • :

    1998 - 2001

      More details

  • :

    1998

      More details

  • :徳島大学 工学部 助手

    1998

      More details

  • :

    1997 - 1998

      More details

  • :東京工業大学 大学院総合理工学研究科 リサーチ・アソシエイト(日本学術振興会 未来開拓学術研究推進事業「生物的適応システム」PD研究員)

    1997 - 1998

      More details

▼display all

Papers

  • Unknown anomaly detection using hidden markov model and areasensing techniques

    Setsuya Kurahashi, Isao Ono

    Tetsu-To-Hagane/Journal of the Iron and Steel Institute of Japan   106 ( 2 )   91 - 99   2020

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:Iron and Steel Institute of Japan  

    DOI: 10.2355/tetsutohagane.TETSU-2019-066

    Scopus

    researchmap

  • A guided local search with iterative ejections of bottleneck operations for the job shop scheduling problem Reviewed

    Yuichi Nagata, Isao Ono

    COMPUTERS & OPERATIONS RESEARCH   90   60 - 71   2018.2

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1016/j.cor.2017.09.017

    Web of Science

    researchmap

  • A grid based simulation environment for agent-based models with vast parameter spaces Reviewed

    Chao Yang, Bin Jiang, Isao Ono, Setsuya Kurahashi, Takao Terano

    CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS   19 ( 1 )   183 - 195   2016.3

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1007/s10586-015-0500-6

    Web of Science

    researchmap

  • A Grid Based Simulation Environment for Parallel Exploring Agent-Based Models with Vast Parameter Space Reviewed

    Chao Yang, Isao Ono, Setsuya Kurahashi, Bin Jiang, Takao Terano

    HUMAN CENTERED COMPUTING, HCC 2014   8944   534 - 548   2015

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-15554-8_44

    Web of Science

    researchmap

  • BS-AWA: A More Scalable Adaptive Weighted Aggregation for Continuous Multiobjective Optimization Reviewed

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    進化計算学会論文誌(Web)   5 ( 1 )   1-15 (J-STAGE) - 15   2014.4

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Evolutionary Computation  

    This paper proposes a more scalable variant of Adaptive Weighted Aggregation (AWA) with respect to the number of objectives in continuous multiobjective optimization. AWA is a scalarization-based multi-start strategy for generating finite points that approximate the entire Pareto set and Pareto front, which is especially focused on many-objective problems (having four or more objectives). In our last study, we discussed a reasonable stopping criterion for AWA, the <em>representing iteration</em>, and analyzed the time and space complexity of AWA when the representing iteration is used as a stopping criterion. Theoretical and empirical results showed that the running time and memory consumption of AWA depends on the number of solutions found in the representing iteration, the <em>representing number</em>. Due to the factorial increase of the representing number for objectives, the applicability of AWA is limited to 16-objective problems. In this study, we therefore redesign two central operations in AWA, the <em>subdivision</em> and the <em>relocation</em>, in order to reduce the representing number. The new subdivision is based on the simplicial complex and its barycentric subdivision and the new relocation is based on the simplicial approximation of a mapping and its range, both of which are well-known notions in topology. We theoretically compare the new AWA, named the <em>barycentric subdivision-based AWA (BS-AWA)</em>, with the old AWA in terms of their representing iteration, representing number and approximate memory consumption to illustrate the improvement of scalability; the result implies that BS-AWA is applicable to over 20-objective problems. Numerical experiments using 2- to 17-objective benchmark problems show that BS-AWA achieves a better coverage of obtained solutions than conventional multi-start descent methods in both the variable and objective spaces. The running time and the solution distribution of BS-AWA are also discussed.

    DOI: 10.11394/tjpnsec.5.1

    J-GLOBAL

    researchmap

  • Improvement of Convergence Properties in Adaptive Weighted Aggregation for Multiobjective Continuous Optimization Reviewed

    Tetsuya Shioda, Yuichi Nagata, Isao Ono

    2014 PROCEEDINGS OF THE SICE ANNUAL CONFERENCE (SICE)   1210 - +   2014

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Random Partial Neighborhood Search for University Course Timetabling Problem Reviewed

    Yuichi Nagata, Isao Ono

    PARALLEL PROBLEM SOLVING FROM NATURE - PPSN XIII   8672   782 - 791   2014

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Improving Estimation Accuracy of Particle Filter by Efficient Interpolation Based on Crossover Reviewed

    Taku Sasaki, Yuichi Nagata, Isao Ono

    2014 PROCEEDINGS OF THE SICE ANNUAL CONFERENCE (SICE)   1216 - +   2014

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • On the Stopping Criterion of Adaptive Weighted Aggregation for Multiobjective Continuous Optimization Reviewed

    濱田直希, 永田裕一, 小林重信, 小野功

    進化計算学会論文誌(Web)   4 ( 1 )   13-27 (J-STAGE) - 27   2013.3

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Evolutionary Computation  

    This paper proposes a reasonable stopping criterion for Adaptive Weighted Aggregation (AWA), which is a scalarization-based multi-start framework developed in our previous study on continuous multiobjective optimization. Our previous study shows that AWA yields good solutions covering the entire Pareto set and front within a small consumption of running time and function evaluation on 2- to 6-objective benchmark problems. The experimental results also indicate, however, that the number of solutions generated by AWA is multiplied every iteration. The rapid increase of solutions requires a careful choice of the stopping criterion: even one iteration of shortage may deteriorate the coverage of solutions into an unsatisfactory level and one of excess gives rise to a significant waste of computational resources. We therefore discuss the minimum iteration that AWA yields an enough solution set to cover the Pareto set and front in the sense that the set contains at least one interior point of each of their non-empty "faces", that is, boundary submanifolds induced from the Pareto sets of subproblems with the same inclusion relation as faces of the simplex. Then, such an iteration, named the <em>representing iteration</em>, is proposed as a stopping criterion for AWA, and the number of solutions found by the representing iteration, named the <em>representing number</em>, is derived to analyze the space complexity of AWA. We also discuss the time complexity of AWA based on numerical experiments. The distribution of obtained solutions and its coverage measure show the usefulness of the proposed stopping criterion.

    DOI: 10.11394/tjpnsec.4.13

    J-GLOBAL

    researchmap

  • An evolutionary algorithm for black-box chance-constrained function optimization

    Kazuyuki Masutomi, Yuichi Nagata, Isao Ono

    Journal of Advanced Computational Intelligence and Intelligent Informatics   17 ( 2 )   272 - 282   2013

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:Fuji Technology Press  

    DOI: 10.20965/jaciii.2013.p0272

    Scopus

    researchmap

  • A Real-Coded Genetic Algorithm Taking Account of the Weighted Mean of the Population Reviewed

    Naotoshi Nakashima, Yuichi Nagata, Isao Ono

    PROCEEDINGS OF THE EIGHTEENTH INTERNATIONAL SYMPOSIUM ON ARTIFICIAL LIFE AND ROBOTICS (AROB 18TH '13)   325 - 328   2013

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • High-Order Sequence Entropies for Measuring Population Diversity in the Traveling Salesman Problem Reviewed

    Yuichi Nagata, Isao Ono

    EVOLUTIONARY COMPUTATION IN COMBINATORIAL OPTIMIZATION (EVOCOP 2013)   7832   179 - +   2013

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Extending Distance-weighted Exponential Natural Evolution Strategy for Function Optimization in Uncertain Environments Reviewed

    Kazuyuki Masutomi, Yuichi Nagata, Isao Ono

    2013 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC)   2122 - 2129   2013

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • A parallel genetic algorithm with edge assembly crossover for 100,000-city scale TSPs

    Kazuma Honda, Yuichi Nagata, Isao Ono

    2013 IEEE Congress on Evolutionary Computation, CEC 2013   1278 - 1285   2013

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1109/CEC.2013.6557712

    Scopus

    researchmap

  • Theoretical Foundation for CMA-ES from Information Geometry Perspective Reviewed

    Youhei Akimoto, Yuichi Nagata, Isao Ono, Shigenobu Kobayashi

    ALGORITHMICA   64 ( 4 )   698 - 716   2012.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1007/s00453-011-9564-8

    Web of Science

    researchmap

  • PATTERN-ORIENTED INVERSE SIMULATION FOR ANALYZING SOCIAL PROBLEMS: FAMILY STRATEGIES IN CIVIL SERVICE EXAMINATION IN IMPERIAL CHINA Reviewed

    Chao Yang, Setsuya Kurahashi, Isao Ono, Takao Terano

    ADVANCES IN COMPLEX SYSTEMS   15 ( 7 )   2012.10

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1142/S0219525912500385

    Web of Science

    researchmap

  • Adaptive Weighted Aggregation: A Multi-Start Framework Taking Account of the Coverage of Solutions for Continuous Multi-Objective Optimization Reviewed

    濱田直希, 永田裕一, 小林重信, 小野功

    進化計算学会論文誌(Web)   3 ( 2 )   31-46 (J-STAGE)   2012.9

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)  

    DOI: 10.11394/tjpnsec.3.31

    J-GLOBAL

    researchmap

  • A New Pareto Frontier Covering Strategy in FS-MOGA for Multi-Objective Function Optimization Reviewed

    Ryo Miyazaki, Naoki Hamada, Yuichi Nagata, Isao Ono

    6TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING AND INTELLIGENT SYSTEMS, AND THE 13TH INTERNATIONAL SYMPOSIUM ON ADVANCED INTELLIGENT SYSTEMS   1888 - 1893   2012

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1109/SCIS-ISIS.2012.6505313

    Web of Science

    researchmap

  • Development of U-Mart System with Plural Brands and Plural Markets

    AKIMOTO Yoshihito, MORI Naoki, ONO Isao, NAKAJIMA Yoshihiro, KITA Hajime, MATSUMOTO Keinosuke

    Transactions of the Society of Instrument and Control Engineers   47 ( 11 )   541 - 548   2011.11

     More details

    Language:Japanese   Publisher:The Society of Instrument and Control Engineers  

    In this paper, we first discuss the notion that artificial market systems should meet the requirements of fidelity, transparency, reproducibility, and traceability. Next, we introduce history of development of the artificial market system named U-Mart system that meet the requirements well, which have been developed by the U-Mart project. We have already developed the U-Mart system called "U-Mart system version 3.0" to solve problems of old U-Mart systems. In version 3.0 system, trading process is modularized and universal market system can be easily introduced.<br/>However, U-Mart system version 3.0 only simulates the single brand futures market. The simulation of the plural brands and plural markets has been required by lot of users. In this paper, we proposed a novel U-Mart system called "U-Mart system version 4.0" to solve this problem of U-Mart system version 3.0. We improve the server system, machine agents and GUI in order to simulate plural brands and plural markets in U-Mart system version 4.0. The effectiveness of the proposed system is confirmed by statistical analysis of results of spot market simulation with random agents.

    DOI: 10.9746/sicetr.47.541

    CiNii Books

    researchmap

    Other Link: https://jlc.jst.go.jp/DN/JALC/00385056579?from=CiNii

  • Adaptive Weighted Aggregation 2: More scalable AWA for multiobjective function optimization. Reviewed

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    Proceedings of the IEEE Congress on Evolutionary Computation, CEC 2011, New Orleans, LA, USA, 5-8 June, 2011   2375 - 2382   2011.6

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    DOI: 10.1109/CEC.2011.5949911

    researchmap

  • On scalability of Adaptive Weighted Aggregation for multiobjective function optimization. Reviewed

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    Proceedings of the IEEE Congress on Evolutionary Computation, CEC 2011, New Orleans, LA, USA, 5-8 June, 2011   669 - 678   2011.6

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    DOI: 10.1109/CEC.2011.5949683

    researchmap

  • Improvement of the Scalability of Adaptive Weighted Aggregation

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    Proceedings of the JSAI 6th Conference on Evolutionary Computation Frontier   82 - 93   2011.3

     More details

    Language:Japanese   Publishing type:Research paper (conference, symposium, etc.)  

    researchmap

  • A New Framework taking account of Multi-funnel Functions for Real-coded Genetic Algorithms Reviewed

    Kento Uemura, Shun-ichi Kinoshita, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    2011 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC)   2091 - 2098   2011

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Analysis on Scalability of AWA with Respect to the Number of Objectives in Many-objective Problems

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    Proceedings of the JPNSEC Symposium on Evolutionary Computation 2010   PDF (6 pages)   2010.12

     More details

    Language:Japanese   Publishing type:Research paper (conference, symposium, etc.)  

    researchmap

  • Adaptive weighted aggregation: A multiobjective function optimization framework taking account of spread and evenness of approximate solutions. Reviewed

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    Proceedings of the IEEE Congress on Evolutionary Computation, CEC 2010, Barcelona, Spain, 18-23 July 2010   1 - 8   2010.7

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    DOI: 10.1109/CEC.2010.5586368

    researchmap

  • Proposal of Multiobjective Function Optimization Framework Taking Account of Spead and Evenness of Approximate Pareto Sets: Adaptive Weighted Aggregation

    Naoki Hamada, Yuichi Nagata, Shigenobu Kobayashi, Isao Ono

    Proceedings of the JSAI 3rd Conference on Evolutionary Computation Frontier   152 - 163   2010.3

     More details

    Language:Japanese   Publishing type:Research paper (conference, symposium, etc.)  

    researchmap

  • A New Real-coded Genetic Algorithm with an Adaptive Mating Selection for UV-landscapes Reviewed

    Oshima,Dan, Miyamae,Atsushi, Nagata,Yuichi, Kobayashi,Shigenobu, Ono,Isao, Sakuma,Jun

    Transactions of the Japanese Society for Artificial Intelligence   25 ( 2 )   290 - 298   2010.1

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Artificial Intelligence  

    The purpose of this paper is to propose a new real-coded genetic algorithm (RCGA) named Networked Genetic Algorithm (NGA) that intends to find multiple optima simultaneously in deceptive globally multimodal landscapes. Most current techniques such as niching for finding multiple optima take into account big valley landscapes or non-deceptive globally multimodal landscapes but not deceptive ones called UV-landscapes. Adaptive Neighboring Search (ANS) is a promising approach for finding multiple optima in UV-landscapes. ANS utilizes a restricted mating scheme with a crossover-like mutation in order to find optima in deceptive globally multimodal landscapes. However, ANS has a fundamental problem that it does not find all the optima simultaneously in many cases. NGA overcomes the problem by an adaptive parent-selection scheme and an improved crossover-like mutation. We show the effectiveness of NGA over ANS in terms of the number of detected optima in a single run on Fletcher and Powell functions as benchmark problems that are known to have multiple optima, ill-scaledness, and UV-landscapes.

    researchmap

  • Development of an artificial market system for analysis of institutional issues in financial markets Reviewed

    Y. Akimoto, N. Mori, I. Ono, Y. Nakajima, H. Sato, H. Matsui, H. Kita, K. Matsumoto

    SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems   121 - 126   2010

     More details

  • Pattern-oriented Inverse Simulation for Agent-based Modeling: An Analysis of Family Strategies Reviewed

    Chao Yang, Setsuya Kurahashi, Isao Ono, Takao Terano

    GECCO-2010 COMPANION PUBLICATION: PROCEEDINGS OF THE 12TH ANNUAL GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE   1801 - 1808   2010

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Analysis of the behavior of MGG and JGG as a selection model for real-coded genetic algorithms Reviewed

    Youhei Akimoto, Yuichi Nagata, Jun Sakuma, Isao Ono, Shigenobu Kobayashi

    Transactions of the Japanese Society for Artificial Intelligence   25 ( 2 )   281 - 289   2010

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)  

    DOI: 10.1527/tjsai.25.281

    Scopus

    researchmap

  • Natural policy gradient methods with parameter-based exploration for control tasks

    Atsushi Miyamae, Yuichi Nagata, Isao Ono, Shigenobu Kobayashi

    Advances in Neural Information Processing Systems 23: 24th Annual Conference on Neural Information Processing Systems 2010, NIPS 2010   2010

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Scopus

    researchmap

  • Theoretical analysis of evolutionary computation on continuously differentiable functions Reviewed

    Youhei Akimoto, Yuichi Nagata, Isao Ono, Shigenobu Kobayashi

    Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10   1401 - 1408   2010

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1145/1830483.1830742

    Scopus

    researchmap

  • Bidirectional Relation between CMA Evolution Strategies and Natural Evolution Strategies Reviewed

    Youhei Akimoto, Yuichi Nagata, Isao Ono, Shigenobu Kobayashi

    PARALLEL PROBLEMS SOLVING FROM NATURE - PPSN XI, PT I   6238   154 - 163   2010

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Globally Multimodal Function Optimization by Real-coded Genetic Algorithms using Traps Reviewed

    Naoya Karatsu, Yuichi Nagata, Isao Ono, Shigenobu Kobayashi

    2010 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC)   2010

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Proposal of Adaptive Weighted Aggregation for Evenly Covering Non-linear Pareto Sets

    Naoki Hamada, Masaharu Tanaka, Jun Sakuma, Shigenobu Kobayashi, Isao Ono

    Proceedings of the JSAI 2nd Conference on Evolutionary Computation Frontier   107 - 112   2009.10

     More details

    Language:Japanese   Publishing type:Research paper (conference, symposium, etc.)  

    researchmap

  • Adaptation of Expansion Rate for Real-coded Crossovers Reviewed

    Youhei, Akimoto, Jun, Sakuma, Isao, Ono, Shigenobu, Kobayashi

    Proceedings of 18th Genetic and Evolutionary Computation Conference (GECCO 2009)   739 - 746   2009.7

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:ACM  

    DOI: 10.1145/1569901.1570004

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/gecco/gecco2009.html#AkimotoSOK09

  • A New Real-coded Genetic Algorithm Using the Adaptive Selection Network for Detecting Multiple Optima Reviewed

    Dan, Oshima, Atushi, Miayamae, Jun, Sakuma, Shigenobu, Kobayashi, Isao, Ono

    Proceedings of IEEE Congress on Evolutionary Computation (CEC) 2009   2009.5

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    researchmap

  • Proposal of Functional-Specialization Multi-Objective Real-Coded Genetic Algorithm: FS-MOGA Reviewed

    濱田直希, 田中雅晴, 佐久間淳, 小林重信, 小野功

    人工知能学会論文誌(Web)   24 ( 1 )   116-126 (J-STAGE)   2009.1

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)  

    DOI: 10.1527/tjsai.24.116

    J-GLOBAL

    researchmap

  • Instance-based Policy Learning by Real-coded Genetic Algorithms and Its Application to Control of Nonholonomic Systems Reviewed

    Miyamae,Atsushi, Sakuma,Jun, Ono,Isao, Kobayashi, Shigenobu

    Transactions of the Japanese Society for Artificial Intelligence   24 ( 1 )   104 - 115   2009.1

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Artificial Intelligence  

    The stabilization control of nonholonomic systems have been extensively studied because it is essential for nonholonomic robot control problems. The difficulty in this problem is that the theoretical derivation of control policy is not necessarily guaranteed achievable. In this paper, we present a reinforcement learning (RL) method with instance-based policy (IBP) representation, in which control policies for this class are optimized with respect to user-defined cost functions. Direct policy search (DPS) is an approach for RL; the policy is represented by parametric models and the model parameters are directly searched by optimization techniques including genetic algorithms (GAs). In IBP representation an instance consists of a state and an action pair; a policy consists of a set of instances. Several DPSs with IBP have been previously proposed. In these methods, sometimes fail to obtain optimal control policies when state-action variables are continuous. In this paper, we present a real-coded GA for DPSs with IBP. Our method is specifically designed for continuous domains. Optimization of IBP has three difficulties; high-dimensionality, epistasis, and multi-modality. Our solution

    researchmap

  • 機能分担 CMA-ES の提案と評価 Reviewed

    秋本, 洋平, 佐久間, 淳, 小野, 功, 小林, 重信

    人工知能学会論文誌   24 ( 1 )   58 - 68   2009.1

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)  

    researchmap

  • Learning of Fugitive Robot Using Optical Information tau Reviewed

    Hiroyuki Fujii, Jun Sakuma, Isao Ono, Shigenobu Kobayashi

    2008 IEEE CONFERENCE ON SOFT COMPUTING IN INDUSTRIAL APPLICATIONS SMCIA/08   20 - 25   2009

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • A Handy Laser Show System for Open Space Entertainment Reviewed

    Toru Takahashi, Miki Namatame, Fusako Kusunoki, Isao Ono, Takao Terano

    ENTERTAINMENT COMPUTING - ICEC 2009   5709   311 - +   2009

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Designing and evaluation of mobile robot using optical information τ-margin

    Hiroyuki Fujii, Yuichi Nagata, Isao Ono, Shigenobu Kobayashi

    Proceedings of the IASTED International Conference on Robotics and Applications   2009

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Scopus

    researchmap

  • Pattern-oriented agent-based simulation for analyzing family strategies in civil service examination in imperial China

    Chao Yang, Toru Takahashi, Takashi Yamada, Setsuya Kurahashi, Isao Ono, Takao Terano

    ICCAS-SICE 2009 - ICROS-SICE International Joint Conference 2009, Proceedings   5121 - 5126   2009

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Scopus

    researchmap

  • A grid-oriented social simulation framework for large scale agent-based modeling

    Chao Yang, Toru Takahashi, Bin Jiang, Takashi Yamada, Isao Ono, Setsuya Kurahashi, Takao Terano

    Conference Proceedings - 6th Conference of the European Social Simulation Association, ESSA 2009   2009

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:The European Social Simulation Association (ESSA)  

    Scopus

    researchmap

  • Proposal of Functional-Specialization Multi-Objective Real-Coded Genetic Algorithm: FS-MOGA

    Hamada Naoki, Tanaka Masaharu, Sakuma Jun, Kobayashi Shigenobu, Ono Isao

    Transactions of the Japanese Society for Artificial Intelligence   24 ( 1 )   116 - 126   2009

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Artificial Intelligence  

    This paper presents a Genetic Algorithm (GA) for multi-objective function optimization. To find a precise and widely-distributed set of solutions in difficult multi-objective function optimization problems which have multimodality and curved Pareto-optimal set, a GA would be required conflicting behaviors in the early stage and the last stage of search. That is, in the early stage of search, GA should perform local-Pareto-optima-overcoming search which aims to overcome local Pareto-optima and converge the population to promising areas in the decision variable space. On the other hand, in the last stage of search, GA should perform Pareto-frontier-covering search which aims to spread the population along the Pareto-optimal set. NSGA-II and SPEA2, the most widely used conventional methods, have problems in local-Pareto-optima-overcoming and Pareto-frontier-covering search. In local-Pareto-optima-overcoming search, their selection pressure is too high to maintain the diversity for overcoming local Pareto-optima. In Pareto-frontier-covering search, their abilities of extrapolation-directed sampling are not enough to spread the population and they cannot sample along the Pareto-optimal set properly. To resolve above problems, the proposed method adaptively switches two search strategies, each of which is specialized for local-Pareto-optima-overcoming and Pareto-frontier-covering search, respectively. We examine the effectiveness of the proposed method using two benchmark problems. The experimental results show that our approach outperforms the conventional methods in terms of both local-Pareto-optima-overcoming and Pareto-frontier-covering search.

    DOI: 10.1527/tjsai.24.116

    DOI: 10.1541/ieejeiss.137.750_references_DOI_MSdCrIWhrMYzityNcA8JeeDPsog

    CiNii Research

    researchmap

    Other Link: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-20700130/

  • Proposal of Multi-Objective Real-Coded Genetic Algorithm Taking Account of Overcoming Local Pareto Solutions and Covering Pareto Frontier

    Naoki Hamada, Masaharu Tanaka, Jun Sakuma, Shigenobu Kobayashi, Isao Ono

    Proceedings of the JPNSEC Symposium on Evolutionary Computation 2008   192 - 197   2008.12

     More details

    Language:Japanese   Publishing type:Research paper (conference, symposium, etc.)  

    researchmap

  • Optimization of Instance-based Policy Based on Real-coded Genetic Algorithms Reviewed

    Atsushi, Miyamae, Jun, Sakuma, Isao, Ono, Shigenobu, Kobayashi

    Proceedings of the 2008 IEEE Conference on Soft Computing in Industrial Applications (SMCia/08)   338-343   2008.9

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    researchmap

  • Functionally specialized CMA-ES: a modification of CMA-ES based on the specialization of the functions of covariance matrix adaptation and step size adaptation Reviewed

    Youhei, Akimoto, Jun, Sakuma, Isao, Ono, Shigenobu, Kobayashi

    Proceedings of Genetic and Evolutionary Computation Conference (GECCO 2008)   479-486   2008.7

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:ACM  

    researchmap

  • Functional-Specialization Multi-Objective Real-Coded Genetic Algorithm: FS-MOGA Reviewed

    Naoki Hamada, Jun Sakuma, Shigenobu Kobayashi, Isao Ono

    PARALLEL PROBLEM SOLVING FROM NATURE - PPSN X, PROCEEDINGS   5199   659 - 669   2008

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • U-Mart System: A Market Simulator for Analyzing and Designing Institutions Reviewed

    Isao ONO, Hiroshi SATO, Naoki MORI, Yoshihiro NAKAJIMA, Hiroyuki MATSUI, Yusuke KOYAMA, Hajime KITA

    Evolutionary and Institutional Economics Review   5 ( 1 )   63 - 79   2008

     More details

    Language:English   Publisher:JAPAN ASSOCIATION FOR EVOLUTIONARY ECONOMICS  

    In this paper, we first discuss the notion that artificial market systems whose purposes are to design institutions for realistic markets should meet the requirements of fidelity, transparency, reproducibility, traceability, and usability. Next, we introduce two artificial market systems named the Itayose U-Mart system and the Zaraba U-Mart system that meet the requirements well, which have been developed by the U-Mart project. Finally, we point out that the U-Mart system is faced with the difficulties of complexities of the system and frequent changes to specification from the viewpoint of software engineering. In order to deal with the difficulties, we employed an object-oriented modeling method to design the U-Mart system and succeeded in constructing the system efficiently.

    DOI: 10.14441/eier.5.63

    researchmap

  • Functional-Specialization Multi-Objective Real-Coded Genetic Algorithm: FS-MOGA Reviewed

    Nanoki Hamada, Jun Sakuma, Shigenobu Kobayashi, Isao Ono

    PARALLEL PROBLEM SOLVING FROM NATURE - PPSN X, PROCEEDINGS   5199   691 - 701   2008

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-540-87700-4_69

    Web of Science

    J-GLOBAL

    researchmap

    Other Link: http://dblp.uni-trier.de/db/conf/ppsn/ppsn2008.html#conf/ppsn/HamadaSKO08

  • Constraint-Handling Method for Function Optimization : Pareto Descent Repair Operator

    HARADA,Ken, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    Transactions of the Japanese Society for Artificial Intelligence   22 ( 0 )   364 - 374   2007.11

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Artificial Intelligence  

    Function optimization underlies many real-world problems and hence is an important research subject. Most of the existing optimization methods were developed to solve primarily unconstrained problems. Since real-world problems are often constrained, appropriate handling of constraints is necessary in order to use the optimization methods. In particular, the performances of some methods such as Genetic Algorithms (GA) can be substantially undermined by ineffective constraint handling. Despite much effort devoted to the studies of constraint-handling methods, it has been reported that each of them has certain limitations. Hence, further studies for designing more effective constraint-handling methods are needed. For this reason, we investigated the guidelines for a method to effectively handle constraints. The guidelines are that the method 1) takes the approach of repair operators, 2) monotonically decreases both the number of violated constraints and constraint violations, and 3) searches over the boundaries of violated constraints. Based on these guidelines, we designed a new constraint-handling method Pareto Descent Repair operator (PDR) in which ideas derived from multi-objectiv

    DOI: 10.1527/tjsai.22.364

    CiNii Books

    researchmap

    Other Link: https://jlc.jst.go.jp/DN/JALC/00292161557?from=CiNii

  • An Experimental Study on Controlling Non-holonomic Systems using Instance-based Policy Learning

    SHIOKAWA,Y, TSUCHIYA,C, SAKUMA,J, ONO,I, KOBAYASHI,S

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   73 - 78   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Even Sampling of Local Pareto-Optimal Solution Curves by Pareto Path Following

    HARADA,Ken, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   377 - 382   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    CiNii Books

    researchmap

  • Hybrid GA for Cardinality-constrained Portfolio Optimization

    TSUCHIYA,Chikao, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   371 - 376   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Generation Alternation Model for Real-coded GA Using Multi-Parent Proposal and Evaluation of Just Generation Gap(JGG)

    AKIMOTO,Youhei, HASADA,Rie, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   341 - 346   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    CiNii Books

    researchmap

  • A Reinitialization Strategy for Real-Coded Genetic Algorithms

    ICHIJIMA,Daijiro, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   335 - 340   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • GA for TSP Considering Global Diversity

    FURUKAWA,Ryo, NAGATA,Yuichi, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   329 - 334   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Evolutionary Design of Zoom Lens Systems by A Real-Coded Genetic Algorithm

    KINOSHITA,Shun-ichi, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   19 ( 0 )   189 - 194   2007.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Constraint-handling method for multi-objective function optimization: Pareto descent repair operator Reviewed

    Ken Harada, Jun Sakuma, Isao Ono, Shigenobu Kobayashi

    EVOLUTIONARY MULTI-CRITERION OPTIMIZATION, PROCEEDINGS   4403   156 - +   2007

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Uniform sampling of local pareto-optimal solution curves by pareto path following and its applications in multi-objective GA.

    Ken Harada, Jun Sakuma, Shigenobu Kobayashi, Isao Ono

    Genetic and Evolutionary Computation Conference(GECCO)   813 - 820   2007

     More details

    Publishing type:Research paper (international conference proceedings)   Publisher:ACM  

    DOI: 10.1145/1276958.1277120

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/gecco/gecco2007.html#HaradaSKO07

  • Saving MGG : Reducing Fitness Evaluations for Real-coded GA/MGG Reviewed

    TANAKA,Masaharu, TSUCHIYA,Chikao, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    Transactions of the Japanese Society for Artificial Intelligence   21 ( 0 )   547 - 555   2006.11

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)   Publisher:The Japanese Society for Artificial Intelligence  

    In this paper, we propose an extension of the Minimal Generation Gap (MGG) to reduce the number of fitness evaluation for the real-coded GAs (RCGA). When MGG is applied to actual engineering problems, for example applied to optimization of design parameters, the fitness calculating time is usually huge because MGG generates many children from one pair of parents and the fitness is calculated by repetitive simulation or analysis. The proposed method called Saving MGG reduces the number of fitness evaluation by estimating the promising degrees of children using individual distribution and fitness information of population, and selecting children based on the promising degree before evaluating the fitness. Experimental results show that RCGA with Saving MGG can provide large reducing effects on 20 or 30 dimensional Sphere functions, Rosenbrock functions, ill-scaled Rosenbrock functions, and Rastrigin function.

    DOI: 10.1527/tjsai.21.547

    CiNii Books

    researchmap

    Other Link: https://jlc.jst.go.jp/DN/JALC/00285086416?from=CiNii

  • Local Search for Multiobjective Optimization : Pareto Descent Method

    HARADA,Ken, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   18 ( 0 )   345 - 350   2006.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Planning and Reinforcement Learning for Grasp and Manipulation

    IWAMI,Kouki, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   18 ( 0 )   143 - 148   2006.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Learning Motion of Deformable Robot Over Rough Terrain

    FUJINO,Tomohiro, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   18 ( 0 )   105 - 110   2006.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Hybridization of Genetic Algorithm with Local Search : Proposal of the Benefits of GA then LS

    HARADA,Ken, SAKUMA,Jun, ONO,Isao, KOBAYASHI,Shigenobu

    自律分散システム・シンポジウム資料 = SICE Symposium on Decentralized Autonomous Systems   18 ( 0 )   351 - 356   2006.1

     More details

    Language:Japanese   Publishing type:Research paper (international conference proceedings)  

    researchmap

  • Instance-Based Policy Search using Binomial Distribution Crossover and Iterated Refreshment. Reviewed

    Chikao Tsuchiya, Kokolo Ikeda, Jun Sakuma, Isao Ono, Shigenobu Kobayashi

    IEEE International Conference on Evolutionary Computation, CEC 2006, part of WCCI 2006, Vancouver, BC, Canada, 16-21 July 2006   378 - 385   2006

  • An effective rule based policy representation and its optimization using inter normal distribution crossover

    Chikao Tsuchiya, Jun Sakuma, Isao Ono, Shigenobu Kobayashi

    Advances in Soft Computing   400 - 411   2005

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:Springer Verlag  

    DOI: 10.1007/3-540-32391-0_47

    Scopus

    researchmap

  • U-Mart system, software for open experiments of artificial market Reviewed

    H Kita, H Sato, N Mori, Ono, I

    2003 IEEE INTERNATIONAL SYMPOSIUM ON COMPUTATIONAL INTELLIGENCE IN ROBOTICS AND AUTOMATION, VOLS I-III, PROCEEDINGS   1328 - 1333   2003

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • U-Mart: An artificial market testbed for economics and multiagent systems Reviewed

    T Terano, Y Shiozawa, H Deguchi, H Kita, H Matsui, H Sato, Ono, I, Y Nakajima

    MEETING THE CHALLENGE OF SOCIAL PROBLEMS VIA AGENT-BASED SIMULATION   53 - 65   2003

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • Theoretical proof of edge search strategy applied to power plant start-up scheduling Reviewed

    Akimoto Kamiya, Kensuke Kawai, Isao Ono, Shigenobu Kobayashi

    IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics   32 ( 3 )   316 - 331   2002.6

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1109/TSMCB.2002.999808

    Scopus

    researchmap

  • Case report on U-Mart experimental system: Competition of software agents and gaming simulation with human agents

    H Sato, H Matsui, Ono, I, H Kita, T Terano, H Deguchi, Y Shiozawa

    AGENT-BASED APPROACHES IN ECONOMIC AND SOCIAL COMPLEX SYSTEMS   72   167 - 178   2002

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Web of Science

    researchmap

  • U-Mart Project: Learning Economic Principles from the Bottom by Both Human and Software Agents. Reviewed

    Hiroshi Sato, Hiroyuki Matsui, Isao Ono, Hajime Kita, Takao Terano, Hiroshi Deguchi, Yoshinori Shiozawa

    New Frontiers in Artificial Intelligence, Joint JSAI 2001 Workshop Post-Proceedings   121 - 131   2001

     More details

    Publisher:Springer  

    DOI: 10.1007/3-540-45548-5_15

    researchmap

  • 実数値 GA のための正規分布交叉の多数の親を用いた拡張法の提案 Reviewed

    喜多 一, 小野 功, 小林 重信

    計測自動制御学会論文集   36 ( 10 )   875 - 883   2000.10

     More details

    Language:Japanese  

    DOI: 10.9746/sicetr1965.36.875

    researchmap

  • An extension of UNDX based on guidelines for designing crossover operators: proposition and evaluation of ENDX Reviewed

    S.Kimura, I.Ono, H.Kira, S.Kobayashi

    Transactions of the Society of Instrument and Control Engineers   36 ( 12 )   1162 - 1171   2000

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)  

    DOI: 10.9746/sicetr1965.36.1162

    J-GLOBAL

    researchmap

  • 実数値GAのための正規分布交叉に関する理論的考察 Reviewed

    喜多 一, 小野 功, 小林 重信

    計測自動制御学会論文集   35 ( 11 )   1333 - 1339   1999.11

     More details

    Language:Japanese  

    DOI: 10.9746/sicetr1965.35.1333

    researchmap

  • The Evolution of Research on Emergent Systems. An Emergent Approach to System Designs.

    TAMAKI Hisashi, ONO Isao, KITAMURA Shinzo

    Journal of The Society of Instrument and Control Engineers   38 ( 10 )   624 - 629   1999.10

     More details

    Language:Japanese   Publisher:The Society of Instrument and Control Engineers  

    DOI: 10.11499/sicejl1962.38.624

    CiNii Books

    researchmap

    Other Link: https://jlc.jst.go.jp/DN/JALC/00064020112?from=CiNii

  • Adaptive-edge search for power plant start-up scheduling Reviewed

    Akimoto Kamiya, Kensuke Kawai, Isao Ono, Shigenobu Kobayashi

    IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews   29 ( 4 )   518 - 530   1999

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1109/5326.798766

    Scopus

    researchmap

  • Designing lens systems taking account of glass selection by real-coded genetic algorithms

    Isao Ono, Yoshihiro Tatsuzawa, Shigenobu Kobayashi, Koji Yoshida

    Proceedings of the IEEE International Conference on Systems, Man and Cybernetics   3   1999

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)  

    Scopus

    researchmap

  • Global and multi-objective optimization for lens design by real-coded genetic algorithms

    Isao Ono, Shigenobu Kobayashi, Koji Yoshida

    Proceedings of SPIE - The International Society for Optical Engineering   3482   110 - 121   1998.9

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE  

    DOI: 10.1117/12.321995

    Scopus

    researchmap

  • A Genetic Algorithm for JSPs Taking Account of Duration between Adjacent Operations Reviewed

    SAKUMA,Jun, ONO,Isao, KOBAYSHI,Shigenobu

    知能システムシンポジウム資料   25 ( 0 )   25 - 30   1998.3

     More details

    Language:Japanese   Publishing type:Research paper (scientific journal)  

    researchmap

  • An Efficient Genetic Algorithm for Reachability Problems. Reviewed

    Keiko Takahashi, Isao Ono, Hiroshi Satoh, Shigenobu Kobayashi

    30th Annual Hawaii International Conference on System Sciences (HICSS-30), 7-10 January 1997, Maui, Hawaii, USA   89 - 98   1997

     More details

    Publisher:IEEE Computer Society  

    DOI: 10.1109/HICSS.1997.663163

    researchmap

  • Genetic algorithm for job-shop scheduling problems using job-based order crossover

    Isao Ono, Masayuki Yamamura, Shigenobu Kobayashi

    Proceedings of the IEEE Conference on Evolutionary Computation   547 - 552   1996

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    Scopus

    researchmap

  • Thermal power plant start-up scheduling with evolutionary computation by using an enforcement operator

    Akimoto Kamiya, Isao Ono, Masayuki Yamamura, Shigenobu Kobayashi

    Proceedings of the IEEE International Conference on Systems, Man and Cybernetics   2   1372 - 1379   1995

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:IEEE  

    Scopus

    researchmap

▼display all

Books

  • Grid Computing in Life Science: First International Life Science Grid Workshop, LSGRID 2004, Revised Selected and Invited Papers(Konagaya, A. and Sato, K. Eds.)

    Springer-Verlag GmbH  2004 

     More details

  • Grid Computing in Life Science: First International Life Science Grid Workshop, LSGRID 2004, Revised Selected and Invited Papers(Konagaya, A. and Sato, K. Eds.)

    Springer-Verlag GmbH  2004 

     More details

  • Advances in Evolutionary Computing (Ghosh, A. and Tsutsui, S., Eds.)

    Springer  2002 

     More details

  • Advances in Evolutionary Computing (Ghosh, A. and Tsutsui, S., Eds.)

    Springer  2002 

     More details

  • 遺伝的アルゴリズム4(北野宏明 編)

    産業図書  2000 

     More details

MISC

  • A New Local Search Method for Multiobjective Black-Box Function Optimization

    2013 ( 108 )   13 - 18   2013.11

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Proposal and Evaluation of a Learning Method for Hearing-Impaired Students through Laser-Show Device in an Open Space

    TAKAHASHI Toru, NAMATAME Miki, KUSUNOKI Fusako, ONO Isao, TERANO Takao

    Journal of Science Education in Japan   34 ( 2 )   117 - 127   2010

     More details

    Language:Japanese   Publisher:Japan Society for Science Education  

    This paper presents a novel learning method for hearing-impaired students in an open space. The method is characterized by the use of a handy laser-show device newly developed by the authors. The learning difficulty of hearing-impaired students comes from the fact that they cannot smoothly interact with each other among students, teachers, and target objects. In a classroom lesson, we are often familiar with the situations, while, in case of an open space, so far, we have had few effective methods. To solve the issues in open space situations, we describe a new learning method, by which we show direct explanations about the target object through a handy laser-show device: Big Fat Wand (BFW). BFW is designed to have capabilities of 1) being carried easily, and 2) presenting characters, images, and animations through authoring tools on a PC. To evaluate the effectiveness, we carried out intensive experiments with hearing-impaired students of Tsukuba University of Technology. The results revealed 1) that a lecturer with no experience or expertise in teaching hearing-impaired students is able to give good lectures with BFW, and 2) that BFW explanations with simple figures about the structure of the target objects are especially educative for hearing-impaired students. We believe that the proposed method will be of use for any other open-space lectures.

    DOI: 10.14935/jssej.34.117

    CiNii Books

    J-GLOBAL

    researchmap

  • 2B2-G2 Investigating an Augmented Reality System for Hearing Impaired People

    TAKAHASHI Toru, NAMATAME Miki, KUSUNOKI Fusako, ONO Isao, TERANO Takao

    Proceedings of the Annual Meeting of Japan Society for Science Education   34 ( 0 )   155 - 156   2010

     More details

    Language:Japanese   Publisher:Japan Society for Science Education  

    DOI: 10.14935/jssep.34.0_155

    CiNii Books

    researchmap

  • 2ZB-3 Analysis of Self-Evaluation for Computer Programming using Project-Based Learning in Graduate School of Informatics

    OKAMOTO Masako, ONO Isao, KIGA Daisuke, TERANO Takao, YAMADA Takashi, KOYAMA Yusuke, MORI Mikihiko, KITA Hajime

    71 ( 4 )   "4 - 583"-"4-584"   2009.3

     More details

    Language:Japanese   Publisher:Information Processing Society of Japan (IPSJ)  

    researchmap

  • Agent-Based Simulation on Women&apos;s Role in a Family Line on Civil Service Examination in Chinese History

    Chao Yang, Setsuya Kurahashi, Keiko Kurahashi, Isao Ono, Takao Terano

    JASSS-THE JOURNAL OF ARTIFICIAL SOCIETIES AND SOCIAL SIMULATION   12 ( 2 )   2009.3

     More details

    Language:English  

    Web of Science

    researchmap

  • 適応的実数値交叉 AREX の提案と評価

    秋本 洋平, 永田 裕一, 佐久間 淳, 小野 功, 小林 重信

    人工知能学会論文誌   24 ( 6 )   446 -- 458 - 458   2009

  • 1B1-D5 The Proposal and the Practice of a Lesson for Hearing-Impaired Students in an Open Space

    TAKAHASHI Toru, NAMATAME Miki, KUSUNOKI Fusako, ONO Isao, TERANO Takao

    Proceedings of the Annual Meeting of Japan Society for Science Education   33 ( 0 )   93 - 94   2009

     More details

    Language:Japanese   Publisher:Japan Society for Science Education  

    DOI: 10.14935/jssep.33.0_93

    CiNii Books

    researchmap

  • Proposal and evaluation of adaptive real-coded crossover AREX

    Youhei Akimoto, Yuichi Nagata, Jun Sakuma, Isao Ono, Shigenobu Kobayashi

    Transactions of the Japanese Society for Artificial Intelligence   24 ( 6 )   446 - 458   2009

     More details

    Language:Japanese   Publisher:Japanese Society for Artificial Intelligence  

    DOI: 10.1527/tjsai.24.446

    Scopus

    researchmap

  • 情報系独立大学院における課題達成型プログラミング教育の分析

    岡本雅子, 小野功, 木賀大介, 寺野隆雄, 山田隆志, 小山友介, 喜多一

    情報処理学会全国大会講演論文集   70th ( 4 )   4.769-4.770   2008.3

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  • Big Fat Wand: Supporting Hering-Impaired Students in a Open Space

    高橋徹, 生田目美紀, 楠房子, 小野功, 寺野隆雄

    人工知能学会全国大会論文集(CD-ROM)   22nd ( 0 )   202 - 202   2008

     More details

  • Lens System Design by A Two Stage GA Solid EMO''

    田中雅晴, 秋本洋平, 佐久間淳, 小野功, 小林重信

    人工知能学会論文誌   23 ( 3 )   193-204 (J-STAGE) - 204   2008

     More details

    Language:Japanese   Publisher:一般社団法人 人工知能学会  

    This paper discusses evolutionary multi-objective optimization (EMO) method for lens system design problems that have properties of global and local multimodality, epistasis among parameters and ill-scaledness. Applying NSGA-II-like EMO to them, it faces some difficulties. To solve them, we present a two stage GA called Solid EMO that consists of a repeated ESO (Evolutionary Single-objective Optimization) and an augmented EMO. The repeated ESO searches seeds of Pareto optimal solutions through solving weighted sum minimization problems repeatedly by a real-coded GA using ISM that deals with global multi-modality well. The augmented EMO, that behaves like a kind of local search by k-nearest neighbor limitation in reproduction and crossover with an ability of explorative search, refines and expands the seeds found by the first stage GA. Solid EMO was applied to three and four element lens system design problems. As a result, the proposed method succeeded in finding highly precise solution sets that consist of well-known types, triplet-type and Lee-type lens systems, in the three-element and four-element lens system design problems, respectively.

    DOI: 10.1527/tjsai.23.193

    J-GLOBAL

    researchmap

  • Using U-Mart System for the lecture of economics

    Yuhsuke Koyama, Ko Ishiyama, Hiroyuki Kaneko, Isao Ono, Hiroyuki Matsui

    AGENT-BASED APPROACHES IN ECONOMIC AND SOCIAL COMPLEX SYSTEMS IV   3   3 - +   2007

     More details

    Language:English  

    Web of Science

    researchmap

  • SLIP : A Sophisticated Learner for Instance-based Policy using Hybrid GA

    TSUCHIYA Chikao, SHIOKAWA Yusuke, IKEDA Kokolo, SAKUMA Jun, ONO Isao, KOBAYASHI Shigenobu

    計測自動制御学会論文集   42 ( 12 )   1344 - 1352   2006.12

     More details

    Language:Japanese   Publisher:計測自動制御学会  

    researchmap

  • Hybridization of Genetic Algorithm with Local Search in Multiobjective Function Optimization : Recommendation of GA then LS

    HARADA Ken, IKEDA Kokolo, SAKUMA Jun, ONO Isao, KOBAYASHI Shigenobu

    Transactions of the Japanese Society for Artificial Intelligence   21   482 - 492   2006.11

     More details

    Language:Japanese   Publisher:The Japanese Society for Artificial Intelligence  

    It is well known that local search (LS) improves the performance of genetic algorithms (GA) in single objective optimization, and it has recently been reported that the hybridization of GA with LS is effective in multiobjective combinatorial optimization as well. In most studies of this kind, LS is applied to the solutions of each generation of GA, which is the scheme called ``GA with LS&#039;&#039; herein. Another scheme, in which LS is applied to the solutions obtained with GA, has also been studied, which is called ``GA then LS&#039;&#039; herein. It seems there is no consensus in the literature as to which scheme is better. The situation in the multibojective function optimization literature is even more unclear since the number of such studies in the field has been small. However, some argue that LS contributes marginally to improving the performance of GA in multiobjective function optimization.&lt;br&gt; This paper, assuming that objective functions are differentiable, reveals the reasons why GA is not necessarily effective in finding solutions of high precision, and hence hybridizing it with LS is indeed effective in multiobjective function optimization. It also suggests that the hybridization scheme which maximally exploits both GA and LS is GA then LS. Experiments confirmed that GA is not suitable for obtaining solutions of high precision, and GA then LS performs better than GA and GA with LS on many benchmark problems.

    DOI: 10.1527/tjsai.21.482

    CiNii Books

    researchmap

  • Local Search for Multiobjective Function Optimization : Pareto Descent Method

    HARADA Ken, SAKUMA Jun, IKEDA Kokolo, ONO Isao, KOBAYASHI Shigenobu

    Transactions of the Japanese Society for Artificial Intelligence   21   350 - 360   2006.11

     More details

    Language:Japanese   Publisher:The Japanese Society for Artificial Intelligence  

    Many real-world problems entail multiple conflicting objectives, which makes multiobjective optimization an important subject. Much attention has been paid to Genetic Algorithm (GA) as a potent multiobjective optimization method, and the effectiveness of its hybridization with local search (LS) has recently been reported in the literature. However, there have been a relatively small number of studies on LS methods for multiobjective function optimization. Although each of the existing LS methods has some strong points, they have respective drawbacks such as high computational cost and inefficiency of improving objective functions. Hence, a more effective and efficient LS method is being sought, which can be used to enhance the performance of the hybridization. &lt;BR&gt; Pareto descent directions are defined in this paper as descent directions to which no other descent directions are superior in improving all objective functions. Moving solutions in such directions is expected to maximally improve all objective functions simultaneously. This paper proposes a new LS method, Pareto Descent Method (PDM), which finds Pareto descent directions and moves solutions in such directions. In the case part or all of them are infeasible, it finds feasible Pareto descent directions or descent directions as necessary and moves solutions in these directions. PDM finds these directions by solving linear programming problems. Thus, it is computationally inexpensive. Experiments have shown that PDM is superior to existing methods.

    DOI: 10.1527/tjsai.21.350

    CiNii Books

    researchmap

  • Inference of Genetic Networks by An Evolutionary Search Taking Account of Simplicity of Network Structures

    清家嘉昭, 小野典彦, 小野功, 中津井雅彦, 岡本正宏

    知能システムシンポジウム資料   33rd   2006

  • 仮想先物市場 U-Mart システムの拡張 Reviewed

    矢和田高大, 小野典彦, 小野功, 中島義裕, 佐藤浩, 森直樹, 松井啓之, 喜多一

    第38 回計測自動制御学会システム工学部会研究会, pp. 71-76, 2006-3.   2006

     More details

    Language:Japanese  

    researchmap

  • An evolutionary algorithm for optimizing functions with UV structures

    Hiroshi Takeichi, Isao Ono, Jun Sakuma, Shigenobu Kobayashi

    2006 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-6, PROCEEDINGS   1296 - +   2006

     More details

  • An artificial market approach to institutional design for thin markets Reviewed

    Hajime Kita, Yoshihiro Nakajima, Isao Ono

    2006 SICE-ICASE INTERNATIONAL JOINT CONFERENCE, VOLS 1-13   2239 - +   2006

     More details

  • グリッド上での遺伝アルゴリズムによるNMR蛋白質立体構造解析

    小野功, 水口尚亮, 中島直敏, 松原彬光, 小野典彦, 中田秀基, 松岡聡, 関口智嗣, 楯真一

    電気学会全国大会講演論文集   2005 ( 3 )   3.S18(11)-3.S18(14)   2005.3

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  • Ninf-1/Ninf-Gを用いたNMR蛋白質立体構造決定のための遺伝アルゴリズムのグリッド化

    小野功, 水口尚亮, 中島直敏, 小野典彦, 中田秀基, 松岡聡, 関口智嗣, 楯真一

    先進的計算基盤システムシンポジウム SACSIS2005   143 - 152   2005

     More details

  • Using U-Mart System for the Lecture of Economics

    Koyama, Y. Ishiyama, K. Matsui, Ono, I

    Proc. 4th Int'l Workshop on Agent-based Approaches in Economics and Social Complex Systems (AESCS'05)   35 - 44   2005

     More details

  • 人工市場プロジェクトU-Mart の教育活動

    喜多一, 谷口和久, 小野功, 松井啓之

    システム/制御/情報   49 ( 7 )   271 - 276   2005

  • A genetic algorithm taking account of substructures for NMR three-dimensional protein structure determination

    N Nakashima, A Matsubara, Ono, I, N Ono, S Tate

    2005 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1-3, PROCEEDINGS   1761 - 1768   2005

     More details

    Language:English  

    Web of Science

    researchmap

  • Ninf-1/Ninf-Gを用いたNMR蛋白質立体構造決定のための遺伝アルゴリズムのグリッド化

    小野功, 水口尚亮, 中島直敏, 小野典彦, 中田秀基, 松岡聡, 関口智嗣, 楯真一

    情報処理学会論文誌:コンピューティングシステム   46 ( SIG12 )   396 - 406   2005

     More details

  • エージェントベースアプローチ 仮想先物市場U‐Martシステムの設計とエージェント・プログラミング教育

    佐藤浩, 小野功, 森直樹

    計測と制御   43 ( 12 )   981 - 986   2004.12

     More details

    Language:Japanese   Publisher:The Society of Instrument and Control Engineers  

    DOI: 10.11499/sicejl1962.43.981

    CiNii Books

    J-GLOBAL

    researchmap

  • U-Mart 仮想経済システムの現状と今後 Reviewed

    喜多一, 小野功, 森直樹, 佐藤浩, 松井啓之, 中島義裕

    JAWS 2004, 合同エージェントワークショップ&シンポジウム2004, pp. 126-131   2004

     More details

    Language:Japanese  

    researchmap

  • 蛋白質立体構造の進化的解析のためのNinf 版並列MGG とその性能評価

    小野功, 今出広明, 中田秀基, 小野典彦, 松岡聡, 関口智嗣, 楯真一

    情報処理学会研究報告 2002-HPC-93(HOKKE2003)   149 - 154   2003

     More details

  • グリッド向け実行環境Jojo を用いた遺伝的アルゴリズムによる蛋白質構造決定

    中田秀基, 中島直敏, 小野功, 松岡聡, 関口智嗣, 小野典彦, 楯真一

    情報処理学会研究報告 2002-HPC-93(HOKKE2003)   155 - 160   2003

     More details

  • 生物的適応システム 〜 進化・学習のアルゴリズムと創発システム論 〜

    小林 重信, 木村 元, 小野 功

    計測と制御 = Journal of the Society of Instrument and Control Engineers   40 ( 10 )   752 - 757   2001.10

     More details

    Language:Japanese   Publisher:計測自動制御学会  

    CiNii Books

    researchmap

  • U-Mart Project. Learning Economic Principles by Agent-Based Simulation.

    佐藤浩, 松井啓之, 小野功, 喜多一, 寺野隆雄, 出口弘, 塩沢由典

    人工知能学会全国大会論文集   15th ( Vol.2 )   3F1.11,1-2 - 245   2001.5

     More details

    Language:Japanese   Publisher:一般社団法人 人工知能学会  

    J-GLOBAL

    researchmap

  • 実数値GAとその応用

    小野功, 山村雅幸, 喜多一

    人工知能学会誌   15 ( 2 )   259 - 266   2000

     More details

  • A Real-Coded Genetic Algorithm for Function Optimization Using the Unimodal Normal Distribution Crossover

    ONO Isao, SATOH Hiroshi, KOBAYASHI Shigenobu, Isao Ono, Hiroshi Satoh, Shigenobu Kobayashi

    Journal of Japanese Society for Artificial Intelligence   14 ( 6 )   1146 - 1155   1999.11

     More details

    Language:Japanese   Publisher:人工知能学会  

    This paper presents a new genetic algorithm (GA) for function optimization, considering epistasis among parameters. When a GA is applied to a function to minimize it, parents are expected to lie on some ponds or along some valleys that are promizing areas because of selection pressure as the search goes on. Especially when the function has epistasis among parameters, it has valleys that are not parallel to coordinate axes. In this case, we believe that a crossover should generate children along the valleys in order to focus the search on such promizing area from a view point of search efficiency. We employ the real number vector as a representation and propose the Unimodal Normal Distribution Crossover (UNDX) taking account of epistasis among parameters. The UNDX generates children near the line segment connecting two parents so that the children lie on the valley where the two parents are when the UNDX is applied to a function with epistasis among parameters. We demonstrate that the UNDX can efficiently optimize various functions including multi-modal ones and ones that have epistasis among parameters by applying he UNDX to some famous benchmark functions.

    researchmap

  • 交叉確率の自己適応によるロバストなGAとその評価

    小野 功, 喜多 一, 小林 重信

    インテリジェント・システム・シンポジウム講演論文集 = FAN Symposium : fuzzy, artificial intelligence, neural networks and computational intelligence   9   21 - 26   1999.10

     More details

    Language:Japanese  

    researchmap

  • 一様交叉と単峰性正規分布交叉の適応的選択に基づくロバストな実数値GA

    小野 功, 喜多 一, 小林 重信

    知能システムシンポジウム資料   26   261 - 266   1999.3

     More details

    Language:Japanese  

    researchmap

  • Multi-parental Extension of the Unimodal Normal Distribution Crossover for Real-coded Genetic Algorithms Reviewed

    Hajime Kita, Isao Ono, Shigenobu Kobayashi

    Proc. CEC99, pp. 1581-1588   1999

     More details

    Language:English  

    DOI: 10.1109/CEC.1999.782672

    researchmap

  • A Robust Real-Coded Genetic Algorithm using Unimodal Normal Distribution Crossover AUgmented by Uniform Crossover: Effects of Self-Adaptation of Crossover Probabilities Reviewed

    Isao Ono, Hajime Kita, Shigenobu Kobayashi

    Proc. GECCO 99, pp. 496-503   1999

     More details

    Language:English  

    researchmap

  • Genetic Algorithms for Scheduling Work Resource Allocation in A Distribution Center

    NUNOYA Satoru, ONO Isao, KOBAYSHI Shigenobu

    25   13 - 18   1998.3

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Theoretical analysis of the unimodal normal distribution crossover for real-coded genetic algorithms Reviewed

    Hajime Kita, Isao Ono, Shigenobu Kobayashi

    Proceedings of the IEEE Conference on Evolutionary Computation, ICEC   529 - 534   1998

     More details

    Language:English  

    researchmap

  • 進化型計算の工学応用

    山村雅幸, 小林重信, 小野功

    第37回計測自動制御学会学術講演会(SICE98)講演論文集   1998

     More details

  • A New Generation Alternation Model of Genetic Algorithms and Its Assessment.

    佐藤浩, 小野功, 小林重信

    人工知能学会誌   12 ( 5 )   734 - 744   1997.9

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  • A Genetic Algorithm based on Sub-sequence Exchange Crossover and GT method for Job-shop Scheduling.

    小野功, 佐藤浩, 小林重信

    電気学会論文誌 C   117-C ( 7 )   888 - 895   1997.7

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  • Emergent search on double circle TSPs using subgour exchange crossover

    M Yamamura, Ono, I, S Kobayashi

    1996 IEEE INTERNATIONAL CONFERENCE ON EVOLUTIONARY COMPUTATION (ICEC '96), PROCEEDINGS OF   535 - 540   1996

     More details

    Language:English  

    Web of Science

    researchmap

▼display all

Presentations

  • Pattern-Oriented Agent-Based Simulation for Analyzing Family Strategies in Civil Service Examination in Imperial China

    ICROS-SICE International Joint Conference 2009  2009 

     More details

  • A Handy Laser Show System for Open Space Entertainment

    8th International Conference on Entertainment Computing (IECE 2009)  2009 

     More details

    Presentation type:Poster presentation  

    researchmap

  • NMR蛋白質立体構造決定のためのα-helixを考慮した部分構造交換交叉の提案

    第32回知能システムシンポジウム  2005 

     More details

  • UV構造を有する関数最適化のための進化アルゴリズムの提案

    計測自動制御学会システム・情報部門学術講演会2005  2005 

     More details

  • 遺伝子ネットワーク推定のための進化アルゴリズムにおける初期集団生成法の提案

    計測自動制御学会システム・情報部門学術講演会2005  2005 

     More details

  • Construction and Operation of the Grid Challenge Testbed

    2006 

     More details

  • グリッドチャレンジテストベッドの構築と運用縲怎Oリチャレテストベッドの作り方縲鰀

    並列/分散/協調処理に関する『高知』サマー・ワークショップ(SWoPP2006)  2006 

     More details

  • 世代交代モデルMGGの並列化とその性能評価

    第32回知能システムシンポジウム  2005 

     More details

  • 遺伝子ネットワーク推定におけるネットワーク構造の網羅的発見のための進化アルゴリズムと性能評価

    第32回知能システムシンポジウム  2005 

     More details

  • グリッド向けGAフレームワークによる逆シミュレーション手法の高速化

    第52回システム制御情報学会研究発表講演会 (SCI'08)  2008 

     More details

  • Big Fat Wand:オープンなスペースでの聴覚障害者教育

    人工知能学会第22回全国大会(JSAI 2008)  2008 

     More details

    Presentation type:Poster presentation  

    researchmap

  • 可搬型レーザデバイスを用いたアクティブ指示装置の聴覚障害者教育への適用と評価

    ヒューマンインタフェースシンポジウム2008  2008 

     More details

  • A Laser Show Device Works in An Open Space for Hearing-Impaired Students

    Second Asia International Conference on Modelling & Simulation (AMS 2008)  2008 

     More details

  • グリッド向け遺伝的アルゴリズムフレームワーク2の提案

    第35回知能システムシンポジウム  2008 

     More details

  • Big Fat Wand:可搬型レーザープロジェクタ

    インタラクション2008  2008 

     More details

    Presentation type:Poster presentation  

    researchmap

  • A Laser Show Device Works in An Open Space for Hearing-Impaired Students

    Second Asia International Conference on Modelling & Simulation (AMS 2008)  2008 

     More details

  • Pattern-Oriented Agent-Based Simulation for Analyzing Family Strategies in Civil Service Examination in Imperial China

    ICROS-SICE International Joint Conference 2009  2009 

     More details

  • A Handy Laser Show System for Open Space Entertainment

    8th International Conference on Entertainment Computing (IECE 2009)  2009 

     More details

    Presentation type:Poster presentation  

    researchmap

  • A Grid-Oriented Social SImulation Framework for Large Scale Agent-Based Modeling

    The Sixth Conference of European Sosial Simulation Association (ESSA 2009)  2009 

     More details

  • プロダクションクラスタ利用型グリッドのためのGAフレームワーク/実行環境の提案

    計測自動制御学会システム・情報j部門学術講演会2009  2009 

     More details

  • A Grid-Oriented Social SImulation Framework for Large Scale Agent-Based Modeling

    The Sixth Conference of European Sosial Simulation Association (ESSA 2009)  2009 

     More details

  • 聴覚障害者のための課外授業デザインの提案と実践

    日本科学教育学会第33回年会  2009 

     More details

▼display all

Awards

  • 計測自動制御学会 論文賞

    2001  

     More details

    Country:Japan

    researchmap

  • 日本光学会 光設計グループ 光奨励賞

    1999  

     More details

    Country:Japan

    researchmap

  • IIZUKA’96 Student Paper Award

    1996  

     More details

  • IIZUKA’96 Student Paper Award

    1996  

     More details

Research Projects

  • 知識発見のための最適化基盤の構築

    Grant number:23K11260  2023.4 - 2026.3

    日本学術振興会  科学研究費助成事業  基盤研究(C)

    小野 功

      More details

    Grant amount:\4420000 ( Direct Cost: \3400000 、 Indirect Cost:\1020000 )

    本年度の主な研究成果は以下の研究1~5のようにまとめられる.研究1では,混合整数ブラックボックス最適化のための自然進化戦略を提案し,整数変数よりも連続変数の目的関数値への寄与が大きいベンチマーク関数において提案手法は既存手法よりも優れた探索性能を示し,その他の既存手法が優れた探索性能示すベンチマーク関数においても既存手法と同等以上の探索性能を示すことを確認した.研究2では,CMA-ESのための学習率適応メカニズムを提案し,集団サイズとして推奨値を用いた提案手法が,学習率チューニングを必要とすることなく,多峰性のベンチマーク関数およびノイズを含むベンチマーク関数において良好な探索性能を示すことを確認した.研究3では,自然進化戦略に基づくUV構造を有する大域的多峰性ブラックボックス関数最適化のためのニッチング手法を提案し,UV構造を有する大域的多峰性のベンチマーク関数において,目的関数のすべての最適解と有力局所解のうち発見できた最適解と有力局所解の割合という観点と目的関数の全ての最適解と有力局所解を発見することができた試行の割合という観点から,提案手法は既存手法よりも優れた探索性能を示すことを確認した.研究4では,騙し構造を考慮した離散ブラックボックス関数最適化のためのVAE-EDAを提案し,強い騙し構造を有するベンチマーク問題において,提案手法は既存手法よりも優れた探索性能を示し,弱い騙し構造を有するベンチマーク問題および騙し構造を有さないベンチマーク問題において,提案手法は既存手法と同等の探索性能を示すことを確認した.研究5では,集団の多様性維持と探索空間の削減に着目したAutoML-Zero手法を提案し,線形回帰アルゴリズムを探索するベンチマーク問題において,最適解発見確率および最適解発見までの評価回数の観点で提案手法が既存手法よりも優れた探索性能を示すことを確認した.

    researchmap

  • Research on multi-layered social models for social risk resilience in smart cities

    Grant number:21H01561  2021.4 - 2026.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

      More details

    Grant amount:\17160000 ( Direct Cost: \13200000 、 Indirect Cost:\3960000 )

    researchmap

  • Research on multi-layered social models for social risk resilience in smart cities

    Grant number:23K21012  2021.4 - 2026.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

      More details

    Grant amount:\17160000 ( Direct Cost: \13200000 、 Indirect Cost:\3960000 )

    researchmap

  • 大域的多峰性探索空間における未知解探索アルゴリズムの深化

    Grant number:20K11986  2020.4 - 2023.3

    日本学術振興会  科学研究費助成事業  基盤研究(C)

    小野 功

      More details

    Grant amount:\4160000 ( Direct Cost: \3200000 、 Indirect Cost:\960000 )

    本年度の主な研究成果は以下の研究1~4のようにまとめられる.研究1では,目的関数の最適解だけでなく多数の局所解を求めるためのニッチング手法として最も優れた手法の1つであるHillVallEAを大域的多峰性関数に適用した際の問題点として,1) 高次元空間において大谷を囲う初期分布を作れず,有力局所解の発見性能が劣化する問題,2) 各大谷が多峰性である関数において探索性能が劣化する問題,3) 同じ大谷を重複して探索してしまい探索効率が悪化する問題を指摘し,これらの問題点に対処した手法を提案した.ベンチマーク問題と4枚組標準レンズ系設計問題への適用を通じて提案手法の有効性を確認した.研究2では,与えられた初期分布が最適解を覆っていなく,初期分布の形状が関数景観の概形と異なるときに評価回数が増加するという自然進化戦略DX-NES-ICの問題点に対処した手法を提案した.ベンチマーク問題を用いた数値実験により,提案手法の有効性を確認した.研究3では,変数間依存関係をもつ離散ブラックボックス関数最適化問題における優れた手法の1つであるBOAの問題点として,変数間依存関係が比較的少ない問題においてベイジアンネットワーク構築時に擬陽性の依存関係を多く検出してしまい,探索性能が劣化する問題を指摘し,それに対処した手法を提案した.ベンチマーク問題への適用を通じて提案手法の有効性を確認した.研究4では,時間枠制約付き配送計画問題において最も優れた手法の1つであるEAMAの問題点として,1) 交叉を行った後の解の種類が少ない問題,2) 集団内の個体の多様性が低下する問題,3) 顧客割当の考慮が不十分である問題を指摘し,これらの問題に対処することを目的とした手法を提案した.ベンチマーク問題への適用を通じて,提案手法の有効性を確認した.

    researchmap

  • Construction of a large scale optimization platform for searching for unkown solutions

    Grant number:17K00335  2017.4 - 2020.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    Ono Isao

      More details

    Grant amount:\4550000 ( Direct Cost: \3500000 、 Indirect Cost:\1050000 )

    In this study, we proposed new search methods that can find better solutions more efficiently than conventional methods for difficult black-box function optimization problems with globally-multimodality, epistasis among parameters, ill-conditionality and implicit constraints. We demonstrated that the proposed methods showed better performance than conventional ones on multiple benchmark problems and real-world applications. In addition, we proposed efficient optimization methods for large-scale traveling salesman problems, estimation methods of differential equation systems for time-series data modeling and simultaneous estimation methods of states and parameters for sequential state estimation problems, and showed that the proposed methods outperformed conventional ones.

    researchmap

  • Proposal of Next Generation Financial System with Market Stability and Liquidity Using Social Simulation

    Grant number:15K01188  2015.4 - 2018.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    MATSUI HIROYUKI, KITA Hajime, TANIGUCHI Kazuhisa, NAKAJIMA Yoshihiro

      More details

    Grant amount:\4680000 ( Direct Cost: \3600000 、 Indirect Cost:\1080000 )

    In this research, 1)we have reproduced the High Frequency Trading finance market using the artificial market system U-Mart System Ver.4. We also built an evaluation system for conducting market simulation. 2)We developed the U-Mart toolkit that is a general-purpose market system and can analyze and evaluate the next generation financial system. 3)It has made the toolkit possible to reproduce and analyze the systemic risk crash mechanism.
    However, we could not produce sufficient results for simulation experiments on the new generation of financial systems that actively utilized the U - Mart Toolkit. Therefore, we plan to continue research further in the future.

    researchmap

  • Constructing Search Algorithms Taking Account of Global-Multimodality and Many-Objectiveness

    Grant number:26330272  2014.4 - 2017.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    Ono Isao

      More details

    Grant amount:\4680000 ( Direct Cost: \3600000 、 Indirect Cost:\1080000 )

    In this study, we proposed new search methods for efficiently finding good approximate solutions in globally-multimodal optimization problems and new ones for efficiently searching for good approximate solution sets in many-objective optimization problems. From the viewpoint of searching for good approximate solutions in the globally-multimodal search space, we proposed new methods to discover big-valleys with the optimal solutions and to search for the best solution in each big-valley. We showed that the proposed methods showed better performance than conventional ones on benchmark problems and real-world ones. From the viewpoint of searching for good approximate solution sets in many-objective optimization problems, we proposed new multi-start search methods based on scalarization that are excellent in terms of the coverage. We demonstrated that the proposed methods outperformed conventional ones on benchmark problems.

    researchmap

  • Constructing Search Algorithms Taking Account of Global Multimodality

    Grant number:23500273  2011 - 2013

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    ONO Isao

      More details

    Grant amount:\5200000 ( Direct Cost: \4000000 、 Indirect Cost:\1200000 )

    In this study, we proposed new search algorithms to efficiently find good solutions in globally multimodal space. It is required that, in the globally multimodal space, search algorithms efficiently find promising big valleys and efficiently search the best solution in each big valley. In order to achieve this requirement, we proposed a framework for finding big valleys and several real-coded evolutionary algorithms for searching in a big valley. The proposed framework iteratively executes real-coded evolutionary algorithms and efficiently find new big valleys by using a history of search regions. We confirmed that the proposed framework with a real-coded genetic algorithm called AREX/JGG succeeded in finding the optima that the conventional methods failed to find on globally multimodal benchmark functions. We also confirmed that the proposed real-coded evolutionary algorithms for searching in a big valley outperformed state-of-the-art algorithms on well-known benchmark functions.

    researchmap

  • The analysis of the systemic risk in the financial markets with large-scale artificial market simulation

    Grant number:23510167  2011 - 2013

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    MATSUI HIROYUKI, ONO Isao, MORI Naoki, KITA Hajime, NAKAJIMA Yoshihiro, TANIGUCHI Kazuhisa

      More details

    Grant amount:\5590000 ( Direct Cost: \4300000 、 Indirect Cost:\1290000 )

    In this research, we aimed to clarify the characteristics of the systemic risk in the financial markets through the use of an artificial market system. Their results are as follows.(1)We have developed the U-Mart system Ver.4 that implements the financial system and market system of reality. (2)We have analyzed the structure of systemic risk in the financial markets, then have developed an experimental platform of artificial market. (3)By performing a simulation analysis using artificial market, and were characterized as theoretically optimal order
    However, it was not possible to give a satisfactory results for the large-scale simulation experiments that had been planned. The large-scale simulation experiments are an issue in the future.

    researchmap

  • Experimental Analysis of the price formation and the market character

    Grant number:22530191  2010 - 2012

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    TANIGUCHI Kazuhisa, NAKAJIMA Yoshihiro, KITA Hajime, ONO Isao, MORI Naoki

      More details

    Grant amount:\1950000 ( Direct Cost: \1500000 、 Indirect Cost:\450000 )

    Even if the buy or sell order prices are the same, the different executed prices are found in a certain market. This study shows the process of the market prices transformation which means the order prices are how to realize executed prices by using the artificial market experiments. The executed prices affect the trader's psychological conditions and the performance of whole market is changed as a result, and the diversities of market trades are also important factor in order to activate the market trades.

    researchmap

  • シミュレーションベースモデリングのための高性能最適化システム基盤の構築

    Grant number:21013018  2009 - 2010

    日本学術振興会  科学研究費助成事業  特定領域研究

    小野 功

      More details

    Grant amount:\5800000 ( Direct Cost: \5800000 )

    本年度の主な研究成果は,以下の2点にまとめられる.
    1)昨年度までに構築したグリッド向けGAフレームワーク/実行環境(Grid-Oriented GA Platform ; GOGAP)を構成する4つのモジュールのうち,グリッド固有のプログラミングの煩雑さを隠蔽して並列プログラミングを容易にするためのJSGF (Java-based Simple Grid Framework)の問題点を克服したJSGF2を提案した.JSGF2では,JSGFの実装効率の問題点を克服するため,Javaプログラマになじみの深いスレッド,オブジェクト,メソッドコールの概念を分散環境に拡張したリモートスレッド,リモートオブジェクト,リモートメソッドコールを用いて並列システムをモデル化するプログラミングモデルを採用している.また,GAにおいて頻繁に用いられるマスタ・ワーカモデルもサポートしている.また,実行効率の問題点を克服するため,リモートメソッドコールはすべて非同期に行われるようになっている.
    2)昨年度までにGOGAP上に構築した社会シミュレーション用フレームワークSOMASの拡張を行った.従来のSOMASでサポートされていた順シミュレーション手法,逆シミュレーション手法,モデル選択手法に加えて,本年度は,パターン指向逆シミュレーション(Patten-oriented Inverse Simulation ; PIS)を提案し,SOMASで利用できるように実装した.PISは,生態学におけるパターン指向モデリング(Pattern-Oriented Modeling ; POM)の考え方を,エージェントシミュレーションにおける逆シミュレーションに導入したものである.SOMAS上に実装されたPISを,歴史シミュレーションに適用することにより,その有用性の検証を行った.

    researchmap

  • Institutional Design and Evaluation of Liquidity Supply in Financial Market Using Participatory Artificial Markets

    Grant number:19300077  2007 - 2009

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

    KITA Hajime, ONO Isao, MORI Naoki, IKEDA Kokolo, MORI Mikihiko, UEHARA Tetsutaro, TANIGUCHI Kazuhisa, MATSUI Hiroyuki, NAKAJIMA Yoshihiro

      More details

    Grant amount:\16900000 ( Direct Cost: \13000000 、 Indirect Cost:\3900000 )

    In security market, it is desirable to trade anytime. Such characteristics are called liquidity of the market. In this study, we discussed the problem of liquidity supply as institutional design problem, and an artificial market approached to the problem is taken. An artificial market system is developed to treat various trading mechanisms such as continuous auction. By gaming using the system, characteristics of the continuous auction market are studied. Further, by agent-based simulation, comparative study of market-maker system and continuous auction is carried out. The results show that the former system is advantageous in liquidity supply in thin markets.

    researchmap

  • シミュレーションベースモデリングのための高性能最適化システム基盤の構築

    Grant number:19024029  2007 - 2008

    日本学術振興会  科学研究費助成事業  特定領域研究

    小野 功, 寺野 隆雄, 岡本 正宏

      More details

    Grant amount:\6000000 ( Direct Cost: \6000000 )

    近年, モデル構築の複雑さの爆発という情報爆発に関する問題に対処するため, 観測データからシミュレーションに基づいて自動的にモデリングを行うシミュレーションベースモデリング(SBM)が注目されている. 現在, SBMのための最適化手法として遺伝アルゴリズム(GA)が有望視されているが, 計算時間の爆発という新たな情報爆発に関する問題に直面している. グリッド計算環境でGAを高速に実行するためのシステム基盤として, 昨年度はGbGA Framework 2の設計, 実装を行った. 本年度は, 社会科学/生命科学の領域専門家がシングルCPU環境からグリッド環境までシームレスに利用可能な環境の構築を目指して, GOGA Framework 2の上にシングルCPU環境用GAライブラリとそのグリッド化モジュールを構築し, 実アプリケーションへの適用を行った.
    シングルCPU環境用GAライブラリは, Javaベースのオブジェクト指向フレームワークであり, 非常に拡張性に富んだ柔軟性の高いライブラリとなっている. 標準的な交叉や世代交代モデルを提供しているが, 利用者は, 交叉, 世代交代モデル, 問題などの独自に開発したモジュールを組み込むことにより, 容易にライブラリを拡張することができる. また, 複数のパラメータについて乱数系列を変更して複数試行を行い, 結果を統計処理するといったSBMにおいて典型的なワークフローもサポートしている.
    GAライブラリのグリッド化モジュールは, シングルCPU環境用GAライブラリをグリッド上で並列分散実行するためのモジュールである. 設定ファイルを差し替えるだけで, 任意のGAについて試行, 個体評価のレベルで並列分散化を実現できる. また, 世代交代モデルに依存する形であるが, 世代交代についても並列分散化するためのモジュールも提供している.

    researchmap

  • Reconstruction and Expansion of Real-coded Genetic Algorithms

    Grant number:19300076  2007 - 2008

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

    KOBAYASHI Shigenobu, ONO Isao, SAKUMA Jun

      More details

    Grant amount:\19110000 ( Direct Cost: \14700000 、 Indirect Cost:\4410000 )

    researchmap

  • 実数値遺伝的アルゴリズムのための適切な初期集団生成法に関する研究

    Grant number:17700154  2005 - 2006

    日本学術振興会  科学研究費助成事業  若手研究(B)

    小野 功

      More details

    Grant amount:\3600000 ( Direct Cost: \3600000 )

    遺伝的アルゴリズム(Genetic Algorithm ; GA)による関数最適化への接近においては,実数値GAのUNDX+MGGが,変数間に強い依存関係をもつ多峰性の大谷関数において良好な性能を示している.しかし,UNDX+MGGは,池田らの提唱するUV構造をもつ多峰性の関数においては最適解の探索に失敗するという問題点をもつ.昨年度,本問題を克服するため,UNDX+MGGを複数回実行する過程において,それまでに十分に探索を行った領域を推定し,その領域を探索領域から除外することにより,最終的に間口の狭いV谷に初期集団を生成する方法を提案し,その有効性を確認した.本年度の研究成果は以下のとおりである:
    1.提案手法が効率よくV谷領域を発見するためには,初期集団が分布している大谷の最良解を効率よく発見できることが必要である.しかし,初期集団が分布している領域の大谷が多峰性であり,かつ,大谷の最良解がオフセットしている場合,UNDX+MGGは最良解の探索に失敗する.本研究では,大谷の最良解がオフセットしている場合,集団分布の重心が移動することに着目し,初期集団を再初期化する方法を提案し,その有効性を確認した.
    2.困難な実問題であるズームレンズ設計へ提案手法を適用するための予備的な研究を行った.本研究では,ズームレンズ設計を,レンズ面の曲率,間隔,群間隔を決定変数とし,歪曲,解像度,焦点距離の誤差の重み付和を評価値とする最小化問題としてモデル化した.2群7枚ズームレンズ設計問題にUNDX+MGGを適用したところ,集団サイズを大きくしても評価値の悪い同じ型の局所解が発見され,本問題がUV構造を有していることが示唆された.ISMを適用した場合,UNDX+MGGよりは良好な解が得られたものの専門家による特許解の発見には至らなかった.現在,提案手法の適用を行っているところである.

    researchmap

  • 仮想先物市場(U-Martシステム)を用いた経済システムの制度創発に関する研究

    Grant number:16016274  2004 - 2005

    日本学術振興会  科学研究費助成事業  特定領域研究

    塩澤 由典, 村上 晴美, 橋本 文彦, 中島 義裕, 谷口 和久, 小野 功

      More details

    Grant amount:\12500000 ( Direct Cost: \12500000 )

    研究成果としては、大きく分けて共通テストベッドの開発と提供、ヒューマン・エージェントの行動分析、マシン・エージェントによる研究の3つに分けられる。人工先物市場U-Martシステムは、ヒューマン・エージェントとマシン・エージェントが混在するシミュレータという特徴を持っている。GUIやログ形成などを統一する事で、ヒューマン・エージェントが参加するリアルタイムの実験と、計算機による加速実験の両方をシームレスに実行できるシステムである。このシステムを開発、公開し国内外の研究機関、教育機関で広く使われている。
    我々は、このシステムを用いてヒューマン・エージェントの行動把握に関する研究を行った。ポジション・コントロールに焦点をあてて学習曲線を調べてみると、6回程度の実験で習熟する事がわかった。また、他人の注文情報(板情報)の有無が取引の成功率に与える影響を調べた。判断する時間が短いと、板情報が活用されない事がわかった。
    マシン・エージェントによる加速実験による成果は2つある。1つは、エージェントの組成が価格変動やエージェントの資産変化にどのような影響を与えるかという問題である。金融市場一般に見られる尖度の高い分布は、エージェントの種類や組み合わせの複雑さが、中程度の時に起きる事がわかった。
    また、約定率を上昇させる事を目的としたマーケット・メーカーのモデルをつくり分析した。どのようなタイプのマーケット・メーカーであれ、注文を継続的に出し続けなければならないという要請を満たすエージェントを加えると取引頻度が増大した。一方で、マーケット・メーカーの収益をみるとスプレッドを固定するタイプのエージェントは価格変動がGARCH型の場合収益をあげられないが、スプレッドをポジションに従って変動するタイプのエージェントは、比較的安定的に収益が得られる事がわかった。

    researchmap

  • 多峰性関数最適化における実数値遺伝的アルゴリズムの探索の効率化に関する研究

    Grant number:15700135  2003 - 2004

    日本学術振興会  科学研究費助成事業  若手研究(B)

    小野 功

      More details

    Grant amount:\3400000 ( Direct Cost: \3400000 )

    遺伝的アルゴリズム(GA)は,多くの局所解をもつ多峰性の探索空間において大域的に良好な解を発見できる強力な近似解法として注目を集めている.GAによる関数最適化への接近においては,特に,UNDX+MGGが,変数間に強い依存関係をもつ多峰性のベンチマーク関数において良好な性能を示している.UNDX+MGGは,全ての変数間に強い依存関係が存在すると仮定して,全ての変数を同時にサンプリングしている.そのため,一部の変数間にしか依存関係をもたない高次元の関数において,UNDX+MGGの探索はかなりの無駄を含んでいると考えられる.特に,多峰性関数においては,同時に探索する変数の数に対して局所解の数が指数関数的に増加するため,問題はさらに深刻になると考えられる.
    本研究では,各変数の依存関係をなるべく正確に推定し,その情報を用いることにより,従来手法であるUNDX+MGGよりも,効率よく探索を行うことができる実数値GAを提案することを目的としている.本年度の研究成果は以下のようにまとめられる:
    ・昨年度,提案したEpistasis-Neighborhood Genetic Hill Climbing (EN-GHC)を困難な大規模実問題へ適用するため,EN-GHCがPCクラスタ上で効率よく計算資源を利用できるような並列モデル(並列EN-GHC)の提案を行った.本モデルは,EN-GHCをマスターワーカー方式に基づき並列化したモデルであり,直接依存関係を持たない変数群の探索を複数のワーカーへ振り分けることにより,並列化効率を向上させている.
    ・並列EN-GHCの並列化効率を調査するため,PCクラスタ上で,昨年度整備したベンチマーク関数へ適用し,良好な結果を得ることに成功した.
    ・現在,EN-GHCとその並列化に関する研究成果に関する論文の投稿へ向けて準備中である.

    researchmap

  • 仮想先物市場(U-Martシステム)を用いた経済システムの制度創発に関する研究

    Grant number:15017276  2003

    日本学術振興会  科学研究費助成事業  特定領域研究

    塩沢 由典, 谷口 和久, 中島 義裕, 村上 晴美, 佐藤 浩, 小野 功

      More details

    Grant amount:\6600000 ( Direct Cost: \6600000 )

    マシン・エージェントとヒューマン・エージェントが同時に参加できる人工先物市場U-Martシステムを用いて、制度設計や制度による市場の間接制御法を確立する事を目的に研究活動を行っている。マシン・エージェントやヒューマン・エージェントを公募して公開実験を開催した。国際大会UMIE2003にはマシン・エージェントが18対参加し、国際会議NAACSOS2003で大会報告及び研究報告を行った。また、マシン・エージェントとヒューマン・エージェントの両方を募集した国内大会U-Mart2003には10体のマシン・エージェントと18人のヒューマン・エージェントが参加し、国際会議ISAGA2003で大会及び研究成果を報告した。これらの大会の実験データを分析し、外部環境(価格時系列)や内部状態(参加エージェントの組成)が市場に与える影響を評価した。昨年の公開実験ではエージェントの組成が異なっても順位の間に相関が見られたが、今年は相関が見られなくなった。また、利用した現物価格の時系列が順位に与える影響に関しても、昨年は上昇/下降というトレンドの方向性が重要な役割を果たしたのに対し、今回の実験ではトレンド変化を捉えられるかどうかがポイントになった。他にもヒューマン・エージェントの行動に関する特徴的な性質を抽出した研究や、学習がすすむにつれて利用する情報や投資方法が異なる事を明らかにした研究なども行われた。これらの研究を促進し、具体的な制度変更の影響を調べるために新システムを構築した。
    制度に関する研究を行う為には、外部環境の整備、内部状態のコントロール、分析方法の確立が必要であるが今年度までの研究で、それらの基礎研究及び基礎データの収集がほぼ完成した。これらの成果を踏まえ2003年末よりマーケット・メーカーに関する本格的な研究が始まっており、基礎モデルの作成と有効性のチェックが進められている。2004年夏を目処にその研究報告を行う予定である。

    researchmap

  • MULTI-AGENT REINFORCEMENT LEARNING WITH NEUROEVOLUTION

    Grant number:14580421  2002 - 2003

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    ONO Norihiko, ONO Isao

      More details

    Grant amount:\3600000 ( Direct Cost: \3600000 )

    Several attempts have been reported to let multiple monolithic reinforcement learning (RL) agents synthesize highly coordinated behavior needed to accomplish their common goal effectively. Most of these straightforward application of RL scale poorly to more complex multi-agent learning problems, because the state space for each RL agent grows exponentially with the number of its partner agents engaged in the joint task.
    To cope with the exponentially large state space in multi-agent RL (MARL), we previously proposed a MARL scheme, based on neural network representation of the decision policy for an agent and its optimization with a real-coded GA, and showed the effectiveness of the scheme through its application to those multi-agent learning problems that can not be solved appropriately using any other conventional MARL scheme, such as the asynchronous multi-agent seesaw balancing problem and the dynamic channel allocation problem in cellular telephone systems.
    However, we can not apply the scheme directly to the design problems of large-scale multi-agent systems (MASs), because the scheme needs a huge amount of computation resources. To remedy the drawback, we propose a hierarchical design scheme of a large-scale MAS, which simply decomposes the whole task of the MAS into its subtasks hierarchically and optimizes each of the subtasks with the above-mentioned MARL scheme. The effectiveness of the design scheme is shown through its application to the RoboCup soccer team design problem where the task of the team is decomposed into the primitive actions by the soccer agents, interaction among the actions and coordination by the agents.

    researchmap

  • 仮想先物市場(U-Martシステム)を用いた経済システムの制度創発に関する研究

    Grant number:14019076  2002

    日本学術振興会  科学研究費助成事業  特定領域研究

    塩澤 由典, 谷口 和久, 北村 泰彦, 中島 義裕, 佐藤 浩, 小野 功

      More details

    Grant amount:\6600000 ( Direct Cost: \6600000 )

    人工先物市場(U-Mart)を開発し、金融市場の制度デザインの研究を行っている。平成14年は、主に(1)サマースクールの開催(2)国際公開実験と国内公開実験の実施と、参加エージェントによる人工市場研究(3)ヒューマンエージェントのによる実験 の3つの活動を行った.
    平成14年7月にサマースクールを開催した。大規模な市場システムを構築する技術を持つ開発スタッフを育てるため、研究分担者を中心に7名が講師となり、大学院生や大学生の20名の受講者に対して教育した。その結果、中心的な開発を行うスタッフが充実し、新サーバーや各種開発ツール、分析ツールが開発された。
    平成14年6月に、カーネギーメロン大学で行われたCASOSカンファレンスの1セッションとして、国際公開実験UMIE2002が開催された。12チーム、48エージェントの応募があった。様々な実験環境の下で、取引実験を行った所、エージェントの順位について、対戦相手の違いよりも外部環境(与えられた時系列)の違いの方が大きな影響を与える事がわかった。
    また、大阪産業大学で大規模なヒューマンエージェントによる実験が行われた.試行回数を増加させるにつれて、参加者の投資スキルが上昇した事がわかった。また、板情報の有無による投資行動の違いについて調べたが、与えられた状況下では明確な違いが発見できなかった。

    researchmap

  • 多峰性関数最適化のための実数値遺伝的アルゴリズムのロバスト化に関する研究

    Grant number:13780287  2001 - 2002

    日本学術振興会  科学研究費助成事業  若手研究(B)

    小野 功

      More details

    Grant amount:\2000000 ( Direct Cost: \2000000 )

    関数最適化問題を解決するための強力な最適化手法として,実数値遺伝的アルゴリズムUNDX+MGGがある.しかし,UNDX+MGGは,多様性を十分に保つために集団サイズを十分に大きくとったとしても,1)探索空間が有界な多峰性関数において最適解が探索空間の境界付近に存在する場合,2)多峰性関数において有望な局所解の存在する大谷の間口が最適解の存在する大谷の間口よりも広い場合に,最適解領域を十分にサンプリングする前に集団を局所解に収束させてしまうという問題点がある.これに対し,昨年度,隔離された複数の部分集団がそれぞれの存在する探索領域を独立に探索を行う「種の棲み分け」の概念を導入した新しい探索モデルを提案し,UNDX+MGGで解決可能な関数に加え,UNDX+MGGでは探索に失敗する上述の性質を持つ関数も解決可能であることを実験により確認した.
    しかし,上述の手法は,部分集団の探索範囲をランダムに生成していたため,探索効率が悪いという問題点があった.また,大規模な実問題への適用において探索時間の観点から問題があり,並列分散化実装が望まれていた.そこで,本年度は,部分集団の探索範囲が広すぎる場合に広い間口の大谷に収束する確率が高くなることに着目し,探索範囲を広い範囲からはじめて,同じ大谷に収束した場合に徐々に探索範囲を縮小していく方法を提案し,その有効性を確認した.また,大規模な実問題への適用の観点から,提案手法の並列分散実装を提案した.本実装は,サーバーへの負荷の集中を回避するために,ピア・ツー・ピア(Peer to Peer ; P2P)モデルを採用し,計算時間を大幅に削減することに成功した.本成果を,計測自動制御学会システムインテグレーション部門講演会にて発表した.

    researchmap

  • 仮想先物市場(U-Martシステム)を用いた経済システムの制度創発に関する研究

    Grant number:13224079  2001

    日本学術振興会  科学研究費助成事業  特定領域研究(C)

    塩沢 由典, 谷口 和久, 北村 泰彦, 中島 義裕, 佐藤 浩, 小野 功

      More details

    今年度はサーバー開発、分析用データベースの構築、ワークショップの開催を中心に活動した。サーバー開発に関する合宿を9月に行い今後の基本方針を固めた。高度な情報提供や大規模シミュレーションに対応するためUDPによる通信を採用した。また、制度設計に関する研究を行う為に必要な機能を洗い出した.基本となる部分は既に実装されているが細かい機能に関しては、むしろ現在進められている研究成果を見てから決定すべきであり、現在は分析ツールの開発に注力している。分析用のデータベースに関しては、大阪証券取引所などから価格情報の提供を受けている。一方これまでU-Mart上で実施したシミュレーション結果をデータベース化し、自由に利用できるような体制を整えた。ワークショップに関しては、5月に大阪市立大学で公開の講習会を行い、人工知能学会(5月)、中央大学商学部(6月)、産業技術総合研究所(7月)、米国のCMU(10月)、甲南大学(12月)、米国ブルッキングス研究所(1月)でデモンストレーションを行った。その結果、実験参加者や教育用の利用予定者が増加した。平成14年7月にCMUで開催されるCASOSでU-Martの国際コンペが開催される事、その際にブルッキングス研究所のアクステル氏が米国内のオーガナイザとして参加する事が決定した。8月にはSICEの創発シンポジウムの中で大規模な公開実験を行い、14チームから39種のプログラムエージェントの参加があった。10月にはFiscoやSimplex等の情報提供や証券取引の教育システムを提供する会社との研究会を開催した。1月の合宿では、宮崎大学の田中氏より経済物理学の講演を聞いた。その後マイクロストラクチャ、効率的市場仮説、人工市場の先行研究に基づく研究プログラムを具体化した。平成14年3月には進化経済学会で専門のセッションを設け研究成果について包括的に報告する。

    researchmap

  • Inference of Genetic Interactions in Large Scale Genetic Network

    Grant number:12208008  2000 - 2004

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research on Priority Areas

    OKAMOTO Masahiro, ONO Isao

      More details

    Grant amount:\97100000 ( Direct Cost: \97100000 )

    The expression profiles of hundreds and thousands of genes on a genomic scale can be measured simultaneously by recent powerful technologies such as DNA microarrays, DNA chips and so forth. These observed data depending on its environment are usually obtained as snapshots, but can be generated as dense time series that indicate the dynamic behavior. The experimentally observed time-course data should contain enormous information about the regulation of genetic networks in vivo. However, since this information is entirely implicit, it requires adequate analytical and computational methods of retrieval and interpretation. This inference problem of genetic networks by using the experimentally observed time-course data is generally referred to as "inverse problem" and can be defined as function optimization of the values of parameters involved in a suitable model representation of genetic network. The key points to solve such an inverse problem are how to set up canonical representation of mathematical modeling of genetic network and how to explore and exploit the values of parameters within immense huge searching space, we had first proposed a novel inferring method of genetic network by combining a dynamic network model called S-system with a computational technique of parameter estimation based on real-coded genetic algorithms (RCGAs). Using S-system modeling and RCGAs with the combination of the UNDX (unimodal normal distribution crossover) and MGG (minmal generation gap), we proposed efficient procedures for the inference of genetic interactions from the experimentally observed time-course data of system components (mRNA). By improving the searching algorithm and by introducing server-client system, we have developed the novel inferring system which can be finding a lots of possibly network candidates that can realize the given experimentally observed time-course data. All of these network candidates can realize the same experimentally observed facts, however, the structures of genetic interactions are different each other. Therefore, we have proposed the analytical method for extracting useful information from many network candidates of. gene expression. In S-system model, the sign of interrelated coefficient shows the kind of interactions such as activation, inhibition, or no relation. The common core interactions are defined by the interactions with sign of which are same among all network candidates of gene expression which inferred based on the same experimentally observed time-course data under the same parameter optimizing conditions. We calculated sensitivity for each interaction included in the network candidates, and compared sensitivity of common core interactions with that of other unique interactions.

    researchmap

  • Multi-agent Reinforcement Learning Based on Compressed Representation of Decision Policies

    Grant number:12680387  2000 - 2001

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    ONO Norihiko, ITO Takuya, ONO Isao

      More details

    Grant amount:\3600000 ( Direct Cost: \3600000 )

    Several attempts have been reported to let multiple monolithic reinforcement learning (RL) agents synthesize highly coordinated behavior needed to accomplish their common goal effectively. Most of these straightforward application of RL scale poorly to more complex multi-agent (MA) learning problems, because the state space for each RL agent grows exponentially with the number of its partner agents engaged in the joint task. To remedy the exponentially large state space in multi-agent RL (MARL), we previously proposed a modular approach and demonstrated its effectiveness through the application to the MA learning problems.
    The results obtained by modular approach to MARL are encouraging, but it still has serious problems. The approach supposes: (i) all the sensory inputs and action outputs for an agent are discrete values, and (ii) all the agents make their decisions totally synchronously at regular time intervals, while such assumption does not hold in real-world multi-agent environments in general.
    We propose yet another MARL framework which can overcome the state space explosion in MARL, based on neural network representation of the decision policy for an agent and its optimization with a real-coded GA, which is applicable to multi-agent domains where individual agents are allowed to receive and output discrete/continuous values and to make their decisions asynchronously. To show the effectiveness of the proposed framework for real-world MARL, we have applied it to the asynchronous multi-agent seesaw balancing problem and the dynamic channel allocation problem in cellular telephone systems. The results are quite encouraging, while those problems can not be solved appropriately using any other conventional MARL frameworks.

    researchmap

  • Synthesis of Coordinated Behavior by Autonomous Agents

    Grant number:10680384  1998 - 1999

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    ONO Norihiko, ITO Takuya, ONO Isao

      More details

    Grant amount:\1000000 ( Direct Cost: \1000000 )

    Several attempts have been reported to let multiple monolithic reinforcement learning (RL) agents synthesize highly coordinated behavior needed to accomplish their common goal effectively. Most of these straightforward application of RL scale poorly to more complex multi-agent (MA) learning problems, because the state space for each RL agent grows exponentially with the number of its partner agents engaged in the joint task. To remedy the exponentially large state space in multi-agent RL (MARL), we previously proposed a modular approach and demonstrated its effectiveness through the application to the MA learning problems.
    The results obtained by modular approach to MARL are encouraging, but it still has a serious problem. The performance of modular RL agents strongly depends on their modular structures, and hence we have to design appropriate structures for the agents. However, it is extremely difficult for us to identify such structures in a top-down manner, because we are not able to correctly predict the performance of a given MA systems, which consists of multiple modular RL agents and accordingly is of substantially complexity with respect to both its structure and its functionality. This means that we have to identify appropriate modular structures for the agents by trial and error. To overcome this problem, we have to establish a framework for automatically synthesizing appropriate modular structures for the agents.
    We suppose that a collection of multiple homogeneous modular RL agents are engaged in a joint task, aimed at the accomplishment of their common goal, and they have the same modular structure in common. We proposed a framework for identifying an appropriate modular structure for the agents, which begins with a randomly generated structure, and attempts to incrementally improve it. A modular structure is represented by a set of a variable number of learning modules, and is evaluated based on the performance of those RL agents employing the structure. The modular structure is improved using a kind of hill-climbing scheme. A set of simple operators is devised, each generating a neighborhood of the current structure.
    To show the effectiveness of the proposed framework, we applied it to a multi-agent learning problem, called the Simulated Dodgeball Game-II and attempted to identify an appropriate modular structure for the attacker agents, each implemented by an independent but homogeneous modular RL architecture. A modular structure is evaluated based on the performance of those attacker agents employing the structure. The results are quite encouraging. Using this framework, for example, we always identified a modular structure which substantially outperforms those manually designed by a human expert.

    researchmap

▼display all